Canvass AI Launches Octane Target Optimization (OTO) Solution for Oil Refiners

3 mins
Canvass AI Launches Octane Target Optimization (OTO) Solution for Oil Refiners

New Solution Saving Refiners $10 Million Annually

TORONTO – January 18, 2024 – Industrial AI software leader Canvass AI today announced availability of a new solution for oil refiners saving millions of dollars in “octane giveaway”. The Canvass AI Octane Target Optimization™ (OTO) solution is implemented at one of the largest North American refineries and is estimated to be saving US $10 million annually in otherwise lost octane giveaway.

Refiners must meet strict regulatory requirements (e.g., Tier 3) during gasoline production for the octane, Reid Vapor Pressure (RVP), and other quality component levels in their product (e.g. sulfur, toxics, oxygenates, etc.). Controlling the levels of octane and RVP as gasoline is produced is complex, and further complicated by additives such as ethanol being blended outside of the refineries causing refiners to blend to conservative targets. While blending optimization can produce gasoline to target octane ranges, the targets are typically set conservatively due to lack of visibility to the final octane values at the pump. This results in revenue losses from octane giveaway. The Canvass AI OTO™ solution predicts both octane from the blending operation as well as final octane after ethanol is added, allowing the refiner to minimize octane giveaway and add millions in revenues.

“Canvass AI OTO stands as a ground-breaking solution empowering refiners to finely tune octane levels, minimizing costly giveaways without the need for significant capital investment. In doing so, it not only enhances operational efficiency but also aids the industry in adhering to government regulations, thereby mitigating the impact on operations and curbing rising costs," stated Humera Malik, CEO of Canvass AI. "In these challenging times, this innovation emerges as a true game-changer for the industry, addressing critical needs and fostering resilience.”

Joe Perino, independent analyst, consultant, and advisor to the oil and gas industry said, “The industry in the US alone is experiencing an estimated average octane giveaway of at least $0.70 per barrel, which amounts to more than a $10 million per year for the refinery in question. AI’s ability to process vast amounts of data quickly and accurately optimizes efficiency, enhances predictive capabilities, and enables automation, all without significant investment.”

The latest OTO solution comes with exciting enhancements including a Prediction Simulator and deep Excel integration.

  • Prediction Simulator: Engineers now have the powerful ability to foresee product outcomes using diverse data sets without impacting the AI model or its training. This feature facilitates 'what-if' analyses, allowing for experimentation with process changes and informed evaluation of potential outcomes.
  • Excel Integration: Recognizing the integral role of Microsoft Excel in the toolkit of process engineers, Canvass AI now seamlessly integrates with Excel. This means users can leverage the robust industrial analytics of Canvass AI while working within the familiar Excel environment. The integration streamlines the process – pulling data from Excel, analyzing it with Canvass AI OTO™ Solution, and seamlessly returning the predictive results to the user's Excel spreadsheet, reducing data wrangling time by an impressive 80%."

For a demo of Canvass AI Solutions visit

About Canvass AI

Canvass AI puts industrial companies in control of their data to achieve operational excellence. Designed to be readily accessible and quick to deploy, companies use Canvass AI’s award-winning patented software for high-performance decision-making, to future-proof operations, and drive net-zero targets. Backed by Alphabet, and Yamaha Motor Ventures, the company is recognized by CB Insights as one of the top technology companies that is advancing manufacturing.

Follow us on X and LinkedIn.

Media and analyst contact:

No items found.
Related Articles